Predicting 2018 Pakistani Election using a Novel Rigged Model

Election results as predicted by this model before elections vs original results.

Introduction

This repository contains code for the election prediction model to predict 2018's general election of Pakistan and won first prize in nation-wide data science competiton. Paper based on this model is accepted in the special issue of Spriger's journal on Big Data and Politics. Find more about this model here.

Dependencies

• scipy 0.18.1
• matplotlib 2.0.0
• pandas 0.19.2
• tqdm 4.28.1
• numpy 1.11.3
• bayesian_optimization 0.6.0
• thesis 0.0.01

How to run the code

cd GE2018
pip install -r requirements.txt
python main.py

Reference

Please cite this.

@article{awais2019leveraging,
  title={Leveraging big data for politics: predicting general election of Pakistan using a novel rigged model},
  author={Awais, Muhammad and Hassan, Saeed-Ul and Ahmed, Ali},
  journal={Journal of Ambient Intelligence and Humanized Computing},
  pages={1--9},
  year={2019},
  publisher={Springer}
}

License

This repository is licensed under the terms of the GNU AGPLv3 license.